Agent-Media UGC Playbook

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides clear instructions for interacting with the Agent-Media API and does not contain any malicious patterns.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials were found; the skill correctly instructs the use of an environment variable ($AGENT_MEDIA_API_KEY) for authentication.
  • [COMMAND_EXECUTION]: No suspicious shell commands or dynamic execution patterns were identified.
  • [DATA_EXFILTRATION]: Network activity is restricted to the legitimate API endpoint api.agent-media.ai required for the skill's functionality.
  • [PROMPT_INJECTION]: Potential surface for indirect prompt injection via user-supplied scripts and media URLs. 1. Ingestion points: script and broll_url parameters in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Network API requests and social media publishing tools. 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 01:25 AM
Security Audit — agent-trust-hub — Agent-Media UGC Playbook